BSD Commit Log Search

   revert wrong change

   dwc_eqos.c: fix grammar in a comment

   Add haud(4), a driver for audio playback using Iris Indigo's
   onboard DSP ("Hollywood Audio"). The same hardware is also
   present on Personal Iris 4D/3x machines, split across the
   mainboard and an optional "Magnum Audio" card (untested).

   The DSP requires firmware, so this driver leverages IRIX's
   hdsp.lod. Since that's proprietary, it's left up to the user
   to install it from an existing IRIX system or its installation
   media. See forthcoming haud(4) manpage for details.

   eqos.4: add details about capabilities

   While here, make some other minor tweaks.

   Fix the MAC address configured for sq(4) on IP12 machines.

   Expansion cards have their own EEPROM with a MAC address at
   offset 250, but my IP12's onboard EEPROM has two MAC addresses
   in the EEPROM at offsets 122 and 250. The former is the system
   MAC address, so check that first and use it if it has the SGI
   OUI before falling back to the 250 offset. Presumably expansion
   cards have nothing at offset 122, but I don't have hardware to
   test.

   awge.4: add definite article

   Remove magic constants set in the cx56 driver's descriptor
   in favour of the hpcreg.h definitions.

   Confusingly, the MI driver apparently swaps the meanings of DO
   and DI in the chip's datasheet.

   Fix EEPROM reading on IP12: DELAY isn't available as early as
   needed, so roll our own using the most pessimistic timings
   (that is, busy loop enough for the fastest pre-ARCS CPU).

   Clean up the EEPROM twiddling code and remove magic constants
   while here.

   Ensure that curlwp is saved and restored across calls into
   pre-ARCS PROM routines. This only matters for console output,
   but do the same for shutdown/restart entrypoints as well.

   While here, preformat strings before invoking the PROM's
   printf routine, since we don't know what its formatting and
   argument limitations are.

   update m4

   Sync m4 with OpenBSD

   regen autoconf stuff.

   doc: Updated math/R-survey to 4.5

   (math/R-survey) Updated 4.4.2 to 4.5

   (pkgsrc)
      - There is a file src/Makevars.makefile, but it does not seem to be used
        (changes on that file does not affect anything) so files/Makevars are
        added for PKG_LIBS+= -lexecinfo

   (upstream)
   4.5  start of first version of multiphase() for arbitrary multiphase sampling
        (experimental, may contain nuts)

        Wilson ('score test') confidence intervals in svyciprop (from Stas Kolenikov)

        svydesign() gives a clearer error message when there are more stages in
        fpc= than in id= (for Stephanie Zimmer)

        svrepdesign(type="ACS") now uses mse=TRUE unless mse=FALSE is explicitly
        given, in which case it warns (for Anthony Damico)


    [75 lines not shown]

   s/IMPOTANT/IMPORTANT/ in comments.

   s/inconsitent/inconsistent/ in comments and panic message.

   fix typos in comments, mainly s/initilize/initialize/.

py-extra-platforms: remove, imported to pkgsrc

   doc: Added sysutils/py-extra-platforms version 13.0.0

   sysutils/Makefile: + py-extra-platforms

   sysutils/py-extra-platforms: import py-extra-platforms-13.0.0

   Extra Platforms detects the runtime architecture, operating system
   (including Linux distribution), shell, terminal, CI environment &
   AI coding agents, and exposes rich, cacheable metadata (version,
   codename, icon, canonical URL). It also groups them into reusable
   families.

py-extra-platforms: remove outdated comment

   igc(4): don't panic when mbuf allocation fails

   Instead, try again later, triggered by a callout.

   Fixes PR 60175.

   ok martin@

   doc: Updated devel/py-filelock to 3.29.3

   py-filelock: update to 3.29.3.

   3.29.3

   🔧 ci(release): publish to PyPI on tag push by @gaborbernat in #557
   validate pid range in _parse_lock_holder by @dxbjavid in #556
   🐛 fix(ci): restore release environment on tag job by @gaborbernat in #559
   🐛 fix(ci): publish from release.yaml on tag push by @gaborbernat in #560

   3.29.2

   open marker reads non-blocking to refuse attacker-placed fifo by @dxbjavid in #549
   🔒 fix(soft): harden stale-lock breaking and self-heal malformed locks by @gaborbernat in #551
   check hostname in is_lock_held_by_us by @dxbjavid in #553

   new file(1)

   bump libmagic

   merge changes between file-5.45 and file-5.48

   Use file_protected for cdf_namecmp rather than static

   Alternative change would be to switch cdf.h the other way.

   Not enough to fix the builds, yet...

   doc: Updated net/xfr to 0.9.16